Computer-implemented method for calibrating a control unit of a motor vehicle
The method addresses inefficiencies in calibrating motor vehicle control units by employing user-selected similarity-based parameter sets and trained data models, enabling rapid and efficient calibration through graphical and predictive architectures.
Patent Information
- Application Number
- DE102024104517
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2024-02-19
- Publication Date
- 2025-07-31
- Estimated Expiration
- 2044-02-19
AI Technical Summary
Existing methods for calibrating control units in motor vehicles are resource-intensive and inefficient, particularly in the context of using artificial intelligence for parameter calibration.
A method involving user input to select parameter sets based on similarity, utilizing a trained data model and predictive embedding architectures to efficiently calibrate control unit parameters, leveraging graphical representations and mathematical operators for comparison.
Facilitates rapid and efficient calibration of control unit parameters by utilizing pre-trained data models, reducing the effort required and enhancing the accuracy through similarity-based comparisons.
Smart Images

Figure 00000000_0000_ABST
Abstract
Claims
[1] Computer-implemented method for calibrating a control unit (1) of a motor vehicle, comprising the following steps: - receiving a user input, wherein the user input comprises an indication of a first parameter set (10'''') of the control unit (1) to be calibrated; - comparing the first parameter set (10'''') with a plurality of second parameter sets (10; 10'; 10''; 10'''; 10''; 10'''''') with regard to a similarity of the first parameter set (10'''') with one of the second parameter sets (10; 10'; 10''; 10'''; 10'''''); - selecting the parameter set from the second parameter sets (10; 10'; 10''; 10'''; 10''''; 10'''''') for which the highest similarity to the first parameter set (10'''') was determined; and - Calibrating the first parameter set (10'''') using a data model (11) and a data set, wherein the data model (11) has been used to calibrate the selected parameter set (10), wherein the data model (11) is trained. [2] Method according to claim 1, characterized by that the user input comprises an indication of a plurality of parameter sets (10; 10'; 10''; 10'''; 10''''; 10'''''; 10'''''') to be calibrated, wherein the parameter sets (10; 10'; 10''; 10'''; 10''; 10'''''; 10'''''') comprise both the first and the second parameter sets. [3] Method according to one of the preceding claims, characterized by that the user input is received as a selection of graphically displayed components (2). [4] Method according to one of the preceding claims, characterized bythat the similarity is an optical similarity of the parameter sets (10; 10'; 10''; 10'''; 10''''; 10''''; 10''''''), wherein the optical similarity is compared using a graphical representation (6; 7; 18; 22; 23) of the respective parameter sets (10; 10'; 10''; 10'''; 10'''; 10'''''; 10''''''). [5] Method according to the preceding claim, characterized by that the comparison of the first parameter set (10'''') with the second parameter sets (10; 10'; 10''; 10'''; 10''''; 10''''') is carried out using a combined predictive embedding architecture (16). [6] Method according to one of the two preceding claims, characterized by that the similarity is determined on the basis of salient features (14; 15; 21) of the graphic representations (6; 7; 18; 22; 23). [7] Method according to one of the preceding claims, characterized by that the similarity is a scalar. [8] Method according to one of the preceding claims, characterized bythat the first parameter set (10'''') and the second parameter sets (10; 10'; 10''; 10'''; 10''''; 10''''') each comprise a functional logic. [9] Method according to one of the preceding claims, characterized by that the first parameter set (10'''') and the second parameter sets (10; 10'; 10''; 10'''; 10''''; 10''''') each comprise input and output parameters. [10] System comprising a digital electronic data memory and a digital electronic processing means, wherein instructions are stored in the data memory which are readable and executable by the processing means, wherein the instructions are designed to cause the processing means, upon execution of the instructions, to carry out a method according to one of the preceding claims.
Citation Information
Patent Citations
Device for determining control unit parameters
DE102009054900A1
Method and apparatus for calibrating a control and regulation device system for controlling and regulating an operating state
DE102020106880A1